© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Qwen2.5-7B-IT
    7. Resid Post - 131k
    9. 15-RESID-POST-AA
    11. 45023
    Prev
    Next
    INDEX
    Explanations

    Code

    np_max-act · gemini-2.0-flash

    strings of hexadecimal digits and numeric identifiers (e.g., commit/file hashes and version IDs).

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-qwen2.5-7b-instruct/resid_post_layer_15/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.15.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     on
    -0.08
     an
    -0.07
    Nb
    -0.07
     من
    -0.07
    Reg
    -0.07
    万象
    -0.07
     over
    -0.07
     Over
    -0.07
     ]]↵
    -0.07
     of
    -0.07
    POSITIVE LOGITS
    posting
    0.09
     bestimm
    0.08
     duplicates
    0.07
     sophisticated
    0.07
     einzel
    0.07
     detached
    0.07
     cortical
    0.07
     damages
    0.06
     dangling
    0.06
     Dickinson
    0.06
    Activations Density 1.144%

    No Known Activations